Abstract The trillions of bacteria that constitutively colonize our intestines (our gut microbiota) produce tens of thousands of unique small molecule metabolites that can potentially affect nearly all aspects of host physiology. However, we only understand the biological functions of a vanishingly small proportion of these chemicals. This reveals a fundamental challenge: how can we identify biologically relevant microbiota metabolites when they are hidden in a sea of other chemicals? Or, in other words, given the presence of tens of thousands of unique unannotated metabolites in the microbiota metabolome, how do we decide which features should be prioritized for in-depth examination and characterization? We propose a revolutionary new approach to these problems. The fundamental concept is to use host sensing of microbiota metabolites as a lens to illuminate the `dark matter' that constitutes the majority of the bioactive microbiota metabolome. Specifically, we hypothesize that the hundreds of G-protein coupled receptors (GPCRs) encoded in the human genome can be used to identify novel bioactive microbiota metabolites from complex mixtures. In proof-of- concept studies using an existing low-throughput GPCR-screening technology to test this approach, we identified metabolite mixtures from individual human gut bacteria that activated GPCRs involved in carcinogenesis, mood regulation, and immunity, as well as `orphan' GPCRs whose natural ligands have evaded discovery for decades. However, because of inherent technical limitations, this methodology can only be applied to a relatively small number of metabolite mixtures that are readily available in large quantities? e.g., metabolites produced by bacterial strains that can be cultured in vitro; thus, it can only capture a small proportion of the overall bioactive microbiota metabolome, which normally results from complex in vivo interactions between multiple microbes, dietary compounds, and the host itself. We thus propose to establish a novel high-throughput technology that exploits the advantages of next-generation sequencing to rapidly screen microbial metabolite mixtures, including precious low-volume samples, against hundreds of GPCRs. We will use this technology to parse thousands of gut microbial metabolomes of varying complexities, ranging from metabolomes produced by individual bacteria all the way to `complete' gut microbiota metabolomes isolated from mice colonized with complex human gut microbial communities. We will then establish a bioassay-guided metabolomic and computational pipeline to simultaneously identify dozens of novel bioactive microbiota metabolites (including ligands for orphan GPCRs), the microbes that produce them, and the receptors through which they mediate their biological effects. These studies will thus (1) provide an unprecedented and transformative view of the bioactive chemicals produced by the microbes that live in and on us, (2) establish a new paradigm for the functional annotation of microbiota metabolites, (3) open up dozens of new avenues for future research, and (4) reveal novel potential therapeutic targets at the host-microbiota interface.